Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 15 May 2019 04:49:54 -0700
From:   Matthew Wilcox <willy@infradead.org>
To:     Michal =?iso-8859-1?Q?Koutn=FD?= <mkoutny@suse.com>
Cc:     khlebnikov@yandex-team.ru, akpm@linux-foundation.org,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org, oleg@redhat.com
Subject: Re: mm: use down_read_killable for locking mmap_sem in
 access_remote_vm
Message-ID: <20190515114954.GB31704@bombadil.infradead.org>
References: <155790847881.2798.7160461383704600177.stgit@buzz>
 <20190515083825.GJ13687@blackbody.suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20190515083825.GJ13687@blackbody.suse.cz>
User-Agent: Mutt/1.9.2 (2017-12-15)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, May 15, 2019 at 10:38:26AM +0200, Michal Koutn? wrote:
> Hi,
> making this holder of mmap_sem killable was for the reasons of /proc/...
> diagnostics was an idea I was pondeering too. However, I think the
> approach of pretending we read 0 bytes is not correct. The API would IMO
> need to be extended to allow pass a result such as EINTR to the end
> caller.
> Why do you think it's safe to return just 0?

_killable_, not _interruptible_.

The return value will never be seen by userspace because it's dead.